DMCA
SVS: Can The Shared Variable Paradigm Exist in Massively Parallel Multiprocessors? (1995)
| Citations: | 1 - 1 self |
BibTeX
@TECHREPORT{Dimitrelos95svs:can,
author = {Dimitris Dimitrelos and Constantine Halatsis},
title = {SVS: Can The Shared Variable Paradigm Exist in Massively Parallel Multiprocessors?},
institution = {},
year = {1995}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
During the last years there have been many efforts aiming towards the elimination of the gap between Shared Memory and Message Passing multiprocessors. Many researchers have addressed the problem by implementing Distributed Shared Memory (DSM) systems. In this paper, we present a solution that is based on the software DSM approach but differentiates from it, as it focuses on keeping shared variables at runtime, without viewing the shared memory as virtual memory, divided in pages. An approach like this can operate complementary to Message Passing environments, taking over the complicated task of process synchronisation, or it can act as an integrated, autonomous system of interprocess communication, if the use of the offered Shared Memory is not heavy. We investigate the viability and the efficiency of such a system in a Massively Parallel environment where non-contention is difficult. SVS (Shared Variable System) is an implementation of this idea, in the form of a runtime library, on ...
Keyphrases
shared variable paradigm exist shared memory massively parallel multiprocessor runtime library distributed shared memory shared variable system autonomous system interprocess communication many effort massively parallel environment virtual memory message passing multiprocessor many researcher process synchronisation last year software dsm approach complicated task
